Proc. Natl. Acad. Sci. USA Vol. 89, pp. 4114-4118, May 1992 Microbiology Mutation in the primer binding site of the type 1 human immunodeficiency genome affects virus production and infectivity T. NAGASHUNMUGAM*, A. VELPANDI*, C. S. GOLDSMITHt, S. R. ZAKIt, V. S. KALYANARAMANt, AND A. SRINIVASAN*§ *The Wistar Institute of Anatomy and Biology, 3601 Spruce Street, Philadelphia, PA 19104; tDivision of Viral and Rickettsial Diseases, Center for Infectious Diseases, Centers for Disease Control, Atlanta, GA 30333; and tAdvanced Bioscience Laboratories, Inc., 5510 Nicholson Lane, Kensington, MD 20805 Communicated by Hilary Koprowski, January 16, 1992

ABSTRACT In an effort to understand the contribution of make use of tryptophan tRNA. Sequence anal- the primer-binding site (PBS) region to human immunodefi- ysis of HIV-1 revealed PBS corresponding to the lysine ciency virus (HIV) replication, we have constructed a mutant tRNA as in other lentiviruses (7). It has also been recently HIV proviral DNA with an alteration in the 5' end of the PBS. demonstrated that HIV RT forms a stable complex with The PBS mutant proviral DNA was characterized by transfec- isoacceptor 3 of lysine tRNA (Lys3 tRNA) (8), as has been tion of the viral DNA into CD4' and non-CD4' target cells. demonstrated for the RT enzyme of avian . The results indicate that mutation in the PBS reduced the level Initiation of viral DNA synthesis is a crucial step in the of viral particles released into the medium of transfected cells replication of retroviruses, and the essential PBS is embed- in comparison to wild-type proviral DNA. The viral particles ded in the complex structure ofRNA (9). A number ofstudies were noninfectious upon transmission to established CD4' cell have reported the importance of the secondary structure of lines and phytohemagglutinin-stimulated peripheral blood the genome at the 5' end in the assembly and infectivity ofthe lymphocytes. Electron microscopic analysis of the transfected virus (10-12). In an effort to understand the structural cells revealed no abnormalities in the structure of the virion requirement of the PBS in reverse transcription and virus directed by the mutant proviral DNA. Also, the protein and morphogenesis, we have introduced an insertional mutation RNA contents of the mutant virions were similar to the wild in the PBS of HIV-1. The results obtained with the PBS type. The quantitation of intracellular viral structural protein mutant imply that the PBS region, in addition to its partici- in the transfected cells, however, indicated that the PBS pation in reverse transcription, is also involved in the assem- mutation may have an effect on the assembly of viral particles bly of viral particles. in addition to completely abolishing reverse transcription of viral RNA into DNA. These results provide evidence that the MATERIALS AND METHODS PBS region of the viral genome has multiple functions in HIV-I Cell Lines. A human rhabdomyosarcoma (RD) cell line replication. (American Type Culture Collection) was maintained as a monolayer culture in Dulbecco's modified Eagle's medium The genome of human immunodeficiency virus type 1 (HIV- supplemented with 10% fetal bovine serum, penicillin (100 1), like other members of the lentivirus family, encodes not units/ml), and L-glutamine (540 pug/ml) at 370C with 5% CO2. only the virion proteins gag, , and common to all CEM x 174 and HUT78 cells were maintained as suspension replication-competent retroviruses but also eight accessory cultures in RPMI 1640 medium; phytohemagglutinin- , some of which are involved in the regulation and stimulated (10 Ag/ml) peripheral blood lymphocytes were morphogenesis of HIV (1-3). Biochemical and structural grown in RPMI 1640 medium containing T-cell growth factor studies have revealed that HIV is about 80-120 nm in (10%). diameter and consists of an inner core (nucleoid) surrounded Construction of the PBS Mutant. The HIV proviral DNA by an outer envelope. The core corresponds to a ribonucleo- plasmid designated pARV was derived from cells infected protein surrounded by the protein. The viral RNA with HIVSF2 (13). The pARV plasmid was cleaved with the genome is in the ribonucleoprotein, and it is composed oftwo restriction enzyme Nar I, and the resultant linear molecule identical positive-sense RNAs with a 5' cap and a 3' poly(A) was purified by agarose gel electrophoresis. The Nar I structure (4). overhang sequence in the DNA was filled in by using Klenow The replication cycle of HIV-1 and other retroviruses polymerase and ligated by using T4 ligase (14). The recom- involves the reverse transcription of virion RNA to viral binant plasmid obtained from the transformed colonies was DNA, which then becomes incorporated into the host cellular checked for the presence of the Nar I cleavage site. The genome. The reverse transcription is mediated by the virion- plasmid (pARV-NAR) lacking the Nar I site was further associated enzyme (RT) (5, 6). As with confirmed by sequence analysis with a Sequenase kit (United all DNA polymerases, RT needs a primer covering a 3' States Biochemical). hydroxyl (OH) group to initiate cDNA synthesis. The in vivo Transfection. RD cells (1 x 106 cells per plate) were split 24 primer for the viral first-strand cDNA synthesis from an RNA hr before transfection, and growth medium was replaced 1-2 template has been shown to be tRNA (6). A region near the hr before the addition of calcium phosphate-precipitated 5' end of the viral RNA genome designated as the primer- DNA (15). Cells were exposed to the precipitate for 8 hr binding site (PBS) is complementary to the 18 nucleotides of followed by a 90-sec glycerol shock (16). For DEAE-dextran the 3' CCA end ofthe specific tRNA primer. Predominantly, transfection, CEM x 174 cells (1 x 107) were washed and many mammalian retroviruses use proline tRNA, and avian Abbreviations: RT, reverse transcriptase; PBS, primer-binding site; The publication costs of this article were defrayed in part by page charge RD, rhabdomyosarcoma; HIV-1, human immunodeficiency virus payment. This article must therefore be hereby marked "advertisement" type 1; RIPA, radioimmunoprecipitation analysis. in accordance with 18 U.S.C. §1734 solely to indicate this fact. §To whom reprint requests should be addressed. 4114 Downloaded by guest on September 24, 2021 Microbiology: Nagashunmugam et al. Proc. Natl. Acad. Sci. USA 89 (1992) 4115

PBS the PBS was confirmed by restriction enzyme and sequence LTR GAG POt ENV LEF analyses. Biological Characterization of HIV-1 Proviral DNA with an Altered PBS. To test the mutant for replication competence, TGGCGCCCGAACAGGGAC pARV and mutant proviral DNAs were directly trans- Nar I wild-type fected into a CD4' cell line designated CEM X 174. A TGGCGcgCCCGAACAGGGAC pARV-NJAR positive result is indicative of a spreading HIV-1 infection in this CD4' cell line. Transfected CEM x 174 cells with FIG. 1. Structure of wild-type and Imutated proviral DNAs used wild-type proviral DNA showed high levels of HIV-1 p24 in this study. The wild-type PBS seque nce is shown. The insertion antigen in the medium, in comparison to the PBS mutant mutation in the PBS region of the muitant proviral DNA (pARV- (Table 1). These results indicate that virus particles directed NAR) is indicated in lowercase letters. LTR, long terminal repeat. by the mutant proviral DNA are unable to initiate a spreading infection. .iumcontainin100 of suspended in 1 ml of serum-free medium containing 100,.ag of Effect of the PBS Mutation on Virus Production. To distin- DEAE-dextran and 10 ,ug ofplasmi( iDNA for 1-2 hr inaPetri guish the effect of the mutation production (synthesis dish. The cells were washed once and were grown in RPMI with 109woserum (17). and assembly of viral particles), we have carried out trans- fection experiments using a transient expression system assay (Coul- HIV Antigen and RT Assay. The HIV antigen(17).en assay (Coul- involving monolayer cells. Human RD cells ideal for ter) was performed according to t these studies, because they have been shown to release high lines to quantitate the amount of herusmanufacTer'assguid procedure was essentially similar to oruat(d8)criTbedR(T asThe levels of viral particles upon transfection as monitored by RT oi and a viral antigen assay (24-26). Transfection of the mutant endogenous RT reaction was carrierd outthatdescribsedn(9In the absence ofofthethe proviral DNA consistently showed a 40-60% reduction in the exogenous template poly(A)-(dT)12 Immunoprecipitation Analysis. i 1D cells102)(2 x trans- release of viral particles in comparison to wild type. The fected with 10 of or 10 IDgof mutant pARV-NAR differences observed in the extracellular viral p24 prompted jig pARV , us agtof mutantol Th to quantitate the intracellular p24 in cells transfected with proviral DNA were labeled 60 hr after glycerol shock. The mutant and wild-type proviral DNA (Table 2). Unlike extra- labeling continued for 12 hr with 8 ml of methionine- and cellular p24 levels, intracellular p24 levels similar in cysteine-deficient serum-free mediu m containinge50r Ciu( Ci both wild-type and mutant proviral DNA-transfected cells = 37 GBq) of Tran35S-label per ml. The cell-free medium and (Table 3). (al ) cells were analyzed by using HIV--positive patient sera and The viral particles released from the transfected cultures normal human sera (21). werre fixed for EM with 2.5% infectivity potential. Since our previous EM. After transfection, cells emixm edtfordEM studies (27) and others (28) have shown that the proviral clone glutaraldehyde, postfixed with osi tetroxide, and em- pARV-derived virus is not able to infect CEM 174 cells bedded in Sectionisum Epon/Araldite. a cell-free virus, we have used the cocultivation method to citrate and uranyl acetate (22). assess the infectivity of the virus. CEM x 174 cells were Nucleic Acid Analysis. The virus particles that sedimented directly added to the proviral DNA-transfected RD cells. The from the medium were lysed and irmpobilized on a nitrogel- cells cocultivated with mutant proviral DNA-transfected RD lulose paper and hybridized to the probe representing full- cells failed to show increase in virus, which indicates the viral DNA For length (23, 24). local]iznglysine 'TCTNATprsen noninfectious nature of the virus derived from the mutant in the virion, the specific oligonu fcleotide 5'TGTGATGG proviral DNA (data not shown). TCTACCGACTGAGCT-3' derive' d from Lys3 tRNA was Biochemical Analysis of Virions. The lack of infectivity of used as a probe (8). the virus derived from the mutant proviral DNA could be due to any of a number of factors, including the absence of virion RESULT'S RNA (29) in the virus particle or the presence of an inappro- priate primer complex that is unable to initiate reverse Construction of an HIV Proviral I)NA with Alteration in the transcription following infection. To determine the presence PBS. To analyze the requirement ()f PBS for of virion RNA, virus particles released from wild-type and viral replication, we initially com; pared the sequences of mutant DNA-transfected cells were pelleted by ultracentrif- number of HIV-1 isolates alreadly available (7). Such a ugation. Based on the p24 antigen values, 25, 50, 75, and 100 comparison revealed few changes in the PBS, and the mu- ng of p24 antigen equivalent virus particles were lysed and tations noted are all present in the 3' end of the PBS. immobilized on a nitrocellulose filter and hybridized to a Inspection of the PBS sequences al[so indicated the presence probe representing the whole viral DNA (Fig. 2). Hybridiza- of recognition sequences for the roestriction enzyme Nar I. tion data showed equal intensities in both wild-type and The unique Nar I cleavage site wars utilized to introduce the mutant virions. Further, hybridization studies with a probe changes in the PBS sequences. HIIV-1 proviral DNA was specific for lysine tRNA also revealed the presence of lysine cleaved with Nar I and treated with Klenow polymerase to fill tRNA packaged in the viral particles. Mutant virus particles in the overhanging sequences. Blunt-end ligation of the DNA were evaluated for the functional RT enzyme molecule. fragment generated a PBS containing two additional bases When the exogenous template primer poly(A)-(dT)12_18 was (Fig. 1). Proviral DNA with this nmutation (pARV-NAR) in used, similar levels of RT activity were noted (Fig. 3A). Table 1. Kinetics of p24 antigen released from transfected CEM x 174 cells p24 antigen released into the medium, pg/ml Proviral DNA 3 days 6 days 9 days 13 days 16 days 19 days 23 days 28 days 32 days pARV 15 21 162 11,500 112,809 119,404 498,019 690,257 563,613 pARV-NAR 15 15 23 33 25 42 11 12 11 CEM x 174 cells (107) were transfected with 10 jig of wild-type pARV or mutant pARV-NAR by the DEAE-dextran method as described in the text. The level of p24 antigen released into the medium was monitored at different intervals (every 3-5 days) after transfection. The cut-off value for this experiment was 7 pg/ml. Downloaded by guest on September 24, 2021 4116 Microbiology: Nagashunmugam et al. Proc. Natl. Acad. Sci. USA 89 (1992) Table 2. Quantitation of the virus particles produced from the A RD cells transfected with wild-type or mutant proviral DNA p24 antigen released into the MOCK medium, pg/ml Proviral DNA Exp. 1 Exp. 2 Exp. 3 pARV * * pARV 47,950 31,441 24,887 pARV-NAR 15,881 16,464 13,840 pARV-Nar * RD cells (106) were transfected with 5 ug of wild-type (pARV) or mutant (pARV-NAR) proviral DNA. The supernatant was assayed 2 3 lt for p24 antigen after 72 hr by an ELISA. The cut-off values were 4 pg/ml for Exp. 1, 5 pg/ml for Exp. 2, and 4 pg/ml for Exp. 3. The B values are from three independent experiments. MOCK However, when RT activity was monitored in the absence of an exogenous template primer, only the wild-type virus showed considerable activity (Fig. 3B). These results clearly pARV * *** indicate that the mutant virus is unable to initiate reverse transcription in an endogenous assay. Analysis of Viral Proteins. The differences in the extracel- pARV-N arr -* lular and intracellular p24 antigen levels prompted us to analyze the viral proteins in the transfected cells. The low level of virus production observed in the mutant DNA- transfected cells could result from a defect in the gag and FIG. 2. Analysis of virion RNA. The viral supernatants obtained envelope protein synthesis and/or processing. The intracel- from RD cells transfected with proviral DNA were quantitated for lular synthesis of viral proteins was analyzed by using p24 antigen. Twenty-five (lane 1), 50 (lane 2), 75 (lane 3), and 100 radioimmunoprecipitation analysis (RIPA). Cells transfected (lane 4) ng of p24 antigen equivalent virus particles were lysed and with the proviral DNA were labeled with [35Sjcysteine and immobilized on nitrocellulose. The filters were hybridized with the [35S]methionine, and RIPA was carried out with both the cell full-length proviral pZ6 probe (A) (23) or with an oligonucleotide extract and the medium. The intracellular and extracellular probe specific for lysine tRNA (B). protein banding patterns in both the wild-type and mutant virus, , spleen necrosis virus, and HIV proviral DNA-transfected cells were indistinguishable (Fig. 4 has indicated that the tRNA primer may, in each case, be and data not shown). positioned within an extensive RNA structure (11, 12). Co- were Analysis ofVirions by EM. Transfected RD cells fixed brinik et al. (10) provided genetic evidence that the secondary 48 hr after glycerol shock to analyze the effect of the PBS mutation on virion morphology by transmission EM. HIV structures in the 5' end of the genome of avian retroviruses particles produced in wild-type and mutant viral DNA- are critical for replication and efficient growth. Sequence transfected cells averaged about 100 nm in diameter and analyses of the PBS in a number of HIV-1 isolates indicate showed identical morphology (Fig. 5). that the PBS is highly conserved, and only minimal changes were noted in the 3' end of the PBS. In this report, we present evidence that the PBS has DISCUSSION multiple functions; it provides a target for anchoring tRNA Although the vital role played by the tRNA bound to the primer and also participates in virion morphogenesis. For genomic RNA as primer has been analyzed (30, 31), studies assessing the effect of the PBS mutation, both wild-type and on the minimal requirement of PBS sequences for binding of primer tRNA and consequences ofalteration in the PBS have A4 :. been limited (32). The 5' noncoding region of retroviral RNA can be subdivided into R, U5, primer-binding, and leader regions. Sequences in these regions are important for reverse transcription, translation, and packaging ofviral RNA and for integration of viral DNA. These regions contain numerous inverted repeat sequences, which can potentially form RNA secondary structures. Analysis of virion RNA by EM re- vealed a stable structure, including U5, a PBS, and leader sequences of and Moloney murine leu- kemia virus. In addition, computer modeling of Rous sar- coma virus, feline leukemia virus, mouse mammary tumor Table 3. Quantitation of intracellular p24 antigen in RD cells transfected with wild-type or mutant proviral DNA Intracellular p24 antigen, pg/ml I'LI Proviral DNA 24 hr 48 hr 72 hr 96 hr 120 hr pARV 27,467 246,605 614,225 252,100 221,070 pARV-NAR 22,950 259,427 757,122 540,945 228,895 RD cells (106) were transfected with 10 jug of wild-type (pARV) or FIG. 3. RT assay. Ten and 20 ng of p24 antigen equivalent virus mutant proviral DNA (pARV-NAR). The cells were collected from particles were analyzed for RT activity in the presence of exogenous the plates at different days after transfection, and the cell pellet was template poly(A)-(dT)12_18 (A) or in the absence of exogenous tem- lysed in 250,u ofdistilled water by freeze-thawing. The cut-off value plate (B) in the reaction. Solid bars, pARV; hatched bars, pARV- in the assay was 5 pg/ml. NAR. Downloaded by guest on September 24, 2021 Microbiology: Nagashunmugam et al. Proc. Natl. Acad. Sci. USA 89 (1992) 4117

1I '2 A PBS z '1 FIG. 4. RIPA. RD cells were transfected with 10 of pARV A 5'UGGCGCCCGAACAGGGAC 3' Genomic RNA Wild Type gP160,1604 jig pARV (lanes 1 and 2) or 10,ug of pARV- 3'ACCGCGGGCUUGUCCCUG 5' 9P120 NAR (lanes 3 and 4) proviral -4 DNA, and cells were labeled with tRNA lysine Tran35S-label (50 iCi/ml), which contains methionine and cysteine, 60 hr after glycerol shock. After a B 5'UGGCG AACAGGGAC3' Mutant P53- _ 12-hr incubation, the cells were pARV-NAR sanwashed with phosphate-buffered 3'ACCGCGGGCUUGUCCCUG 5' saline and iysed in RIPA buffer (0.15 mM NaCl/0.05 mM _ Tris HCI, pH 7.2/1% Triton C 5'UGGC X-100/1% sodium deoxycholate/ GcgCCCGAACAGGGAC3' Mutant 0.1% SDS). Immunoprecipitation pARV-NAR p24-_ was done with normal sera (lanes CGCGGGCUUGUCCCUG 5' 1 and 3) or with HIV-positive sera 3'---AC __ . (lanes 2 and 4). FIG. 6. Interaction of lysine tRNA with wild-type and PBS mutant genomic RNA. (A) Sequence homology between the 3' end mutant proviral DNA were transfected into CD4+ CEM x of the tRNA and the genomic RNA PBS enables successful reverse 174 cells. This experimental system relies on the ability ofthe transcription in wild-type virions. (B and C) Possible models for the virus to initiate a new infection upon release from the interaction of lysine tRNA with the PBS mutant genomic RNA. transfected cells. Generally, a high amount of RT activity or Mutant sequences in the PBS region are indicated by lowercase viral antigen is observed after 10-15 days. The results gen- letters. The arrow indicates the direction of reverse transcription. erated by using this system clearly indicated that the virus released by mutant proviral DNA is noninfectious because to generate HIV upon transfection. These target cells lack there was no spreading infection. CD4 receptors on the cell surface, and this feature enables To characterize the effect of the PBS mutation on virus absolute quantitation of virus released from the transfected production, we have used monolayer cells as target cells for cells. PBS mutant proviral DNA showed a low level of virus transfection of proviral DNA. We and others (24, 25, 27, 28) production in comparison to the wild-type DNA. These have extensively used human RD cells, as a transient system, results are in sharp contrast to the data recently reported by Rhim et al. (32) involving deletion mutational analysis of the PBS. The difference between our data and that of Rhim et al. 4r~~~~~~~~~~~~~~~- (32) may be due to the difference in the cell system used for A£5 transfection studies. In RD cells, HIV has been shown to bud from the cell surface as observed in H9, MOLT-3, MOLT-4, and CEM cells (4). The reduced level of virus production observed could result from a number of events including transcription, RNA stability, translation, and posttransla- tional processing. The quantitation of the viral structural protein p24, in the form of viral particles released into the medium, showed low levels from mutant DNA-transfected cells. However, the levels of intracellular structural protein of the virus remained the same in both the mutant and wild-type DNA transfected cells, indicating a possible defect in any one of the steps associated with the virion assembly. It has been reported that the RT enzyme and nucleocapsid protein participate in the selection and binding of the tRNA primer to the PBS in the genomic RNA. The leader and replication primer tRNA are highly structured, and hybrid- B ; ization seems unlikely without disruption of secondary struc- tures. A role for the nucleocapsid protein in the unwinding of the RNA structure, which promotes a base pairing interaction between complementary sequences, has been reported (8, 9). ~ '~ - In the presence of RT, in addition to nucleocapsid protein, tRNA binds more the PBS. It is that the A efficiently to possible Ir -1 PBS mutation interferes with this process, which results in 4b.. .l A., diminished .'. virion morphogenesis. lb Concomitant to the data observed in CEM x 174 cells, virus released from the monolayer cells was noninfectious in an assay involving cocultivation of CEM x 174 cells with proviral DNA-transfected RD cells. The lack of infectivity of the virus could result from impaired processing of viral proteins. Studies have shown that the cleavage of the env -'1- '"'",' '44X a ii. precursor gpl60 into the gpl2O- heterodimer is critical for infectivity (33). HIV particles containing uncleaved gpl60 due to a mutation are not infectious. Similarly, the HIV- FIG. 5. EM of transfected RD cells. Cells transfected with specific protease has to process the p55aag precursor to render proviral DNA (A, pARV; B, pARV-NAR) were collected 72 hr after the assembled particles infectious (34, 35). To rule out this glycerol shock. The cells were fixed and analyzed by transmission possibility, analysis of viral proteins in the transfected cells electron microscopy. (x 15,500.) was carried out by using sera from HIV-positive individuals. Downloaded by guest on September 24, 2021 4118 Microbiology: Nagashunmugam et al. Proc. Natl. Acad. Sci. USA 89 (1992) The protein profile was similar in both wild-type and PBS 8. Barat, C., LeGrice, S. F. J. & Darlix, J. L. (1991) Nucleic mutant DNA-transfected cells. Acids Res. 19, 751-757. Biochemical analysis of the viral particles directed by the 9. Barat, C., Lullien, V., Schatz, O., Keith, G., Nugeyre, M. T., PBS mutant proviral DNA was carried out to understand the Gruninger-Leitch, F., Barre-Sinoussi, F., LeGrice, S. F. J. & basis of the noninfectious nature of the virus. Hybridization Darlix, J. L. (1989) EMBO J. 8, 3279-3285. analysis of the viral particles utilizing the probes specific for 10. Cobrinik, D., Soskey, L. & Leis, J. (1988) J. Virol. 62, 3622-3630. the HIV-1 genome and the lysine tRNA primer showed that 11. Darlix, J. L., Zuker, M. & Spahr, P.-F. (1982) Nucleic Acids packaging of the viral RNA and the tRNA primer is not Res. 10, 5183-51%. affected. The viral particles derived from the PBS mutant 12. Murti, K. G., Bondurant, M. & Tereba, A. (1981) J. Virol. 37, were also characterized for the presence of functional RT 411-419. enzyme. Although functional RT enzyme was detected by 13. Levy, J. A., Cheng-Mayer, C., Dina, D. & Luciw, P. A. (1986) using an exogenous template primer, endogenous reverse Science 232, 998-1001. transcription with the PBS mutant virus was negligible. These 14. Sambrook, J., Fritsch, E. F. & Maniatis, J. (1989) Molecular results indicated that there is a block at the initiation of Cloning: A Laboratory Manual (Cold Spring Harbor Lab., Cold reverse transcription in the mutant virus. Spring Harbor, NY), 2nd Ed. The PBS mutation analyzed in this study contains a 2-base 15. Frost, E. & Williams, J. (1978) Virology 91, 39-50. insertion, which is located 5 bases downstream of the 5' end 16. Graham, F. W. & Van der Eb, J. (1973) Virology 52, 456-467. of the PBS. Even though there is evidence for the use of 17. Stafford, J. & Queen, C. (1983) Nature (London) 306, 77-79. truncated tRNA primers (36-38) or incorrectly processed 18. Goudsmit, J., Paul, D. A., Lange, M. A., Speelman, H., Vander Noordan, J., Vander Helm, H. J., Wolf, F., Epstein, tRNA (39), the insertion of 2 bases completely eliminated the L. G., Krone, W. J. A., Walters, E. C., Oleske, J. M. & PBS function in reverse transcription. Because of the ho- Coutinho, R. A. (1986) Lancet ii, 177-180. mology, tRNA anneals efficiently to the genomic RNA PBS 19. Adachi, A., Gendelman, H. E., Koenig, S., Folks, T., Willey, and is able to initiate the reverse transcription process in the R., Rabson, A. & Martin, M. A. (1986) J. Virol. 59, 284-291. wild type (Fig. 6A). The lack of endogenous reverse tran- 20. Yoshida, M., Miyoshi, I. & Hinuma, Y. (1982) Proc. Natl. scription in the mutant virions could be attributed to two Acad. Sci. USA 79, 2031-2035. possible mechanisms, as depicted in Fig. 6 B and C. In the 21. Kalyanaraman, V. S., Pal, R., Gallo, R. C. & Sarangadharan, first model (Fig. 6B), the tRNA primer binds to the genomic M. G. (1988) AIDS Res. Hum. Retroviruses 4, 319-329. RNA, leaving nonhomologous mutant sequences in the PBS 22. Palmer, E., Sporborg, C., Harrison, A., Martin, M. L. & as a bulge, which would then lead to successful initiation of Feorino, P. (1985) Arch. Virol. 85, 189-1%. reverse transcription. The results obtained with endogenous 23. Srinivasan, A., Anand, R., York, D., Ranganathan, P., Fe- reverse transcription, however, orino, P., Schochetman, G., Curran, J., Kalyanaraman, V. S., were in favor of the second Luciw, P. A. & Sanchez-Pescador, R. (1987) Gene 52, 71-82. model (Fig. 6C), where binding occurs only at the 3' end of 24. Srinivasan, A., Goldsmith, C. S., York, D., Anand, R., Luciw, PBS with the tRNA molecule. The nonhomology at the5' end P., Schochetman, G., Palmer, E. & Bohan, C. (1988) Arch. of PBS, because of the insertion of bases, leaves the 3' OH Virol. 99, 21-30. termini of the tRNA primer ineffective in priming. Such an 25. Srinivasan, A., York, D., Jannoun-Nasr, R., Kalyanaraman, observation has also been noted with oligonucleotide primers S., Swan, S. D., Benson, J., Bohan, C., Luciw, P. A., Schnoll, in the polymerase chain reaction (40). S., Robinson, R. A., Desai, S. M. & Devare, S. G. (1989) Proc. The exact mechanisms by which alteration in the PBS Natl. Acad. Sci. USA 86, 6388-6392. affects virus production are not known. The observations of 26. Velpandi, A., Monken, C. E. & Srinivasan, A. (1990) J. Virol. increased levels Methods 25, 291-302. of structural protein p24 inside the cells, 27. Velpandi, A., Nagashunmugam, T., Murthy, S., Cartas, M., dissimilar levels of virus released into the cultural medium of Monken, C. & Srinivasan, A. (1991) J. Virol. 65, 4847-4852. the proviral DNA-transfected cells, and the noninfectious 28. Cheng-Mayer, C., Seto, D. & Levy, J. A. (1991) Virology 181, nature of the virus produced by the mutant (pARV-NAR) 288-294. suggest that the PBS mutation possibly interferes with the 29. Gorelick, R. J., Nigida, S. M., Jr., Bess, J. W., Jr., Arthur, processes associated with the virion morphogenesis in addi- L. O., Henderson, L. E. & Rein, A. (1990) J. Virol. 64, tion to abolishing reverse transcription. These results support 3207-3211. the notion that the PBS region of the retroviral genome has 30. Prats, A. C., Sarih, L., Gabus, C., Litvak, S., Kulk, G. & multiple functions in replication (41, 42). Addi- Darlix, J. L. (1988) EMBO J. 7, 1777-1783. 31. Prats, A. C., Honsset, V., deBilly, G., Carmille, F., Prats, H., tionally, the noninfectious HIV-1 resulting from the PBS Rogues, B. & Darlix, J. L. (1991) Nucleic Acids Res. 19, mutation provides an important alternative source of immu- 3533-3541. nization of animals for evaluating vaccine strategies. 32. Rhim, H., Park, J. & Morrow, C. D. (1991) J. Virol. 65, 4555-4564. We thank the Wistar Institute editorial staff for preparation of the 33. McCune, J. M., Rabin, L. B., Feinberg, M. B., Lieberman, manuscript. We gratefully acknowledge P. A. Luciw for providing M., Kosek, J. C., Reyes, G. R. & Weissman, L. (1988) Cell 53, proviral DNA derived from HIVSF2 and J. Hoxie for CEM x 174 55-67. cells. This work was supported by Grants A129306 and A125380 from 34. Kohl, N. E., Emini, E. A., Schleif, W. A., Davis, L. J., Heim- the National Institutes of Health. bach, J. C., Dixon, R. A. F., Scolnick, E. M. & Sigal, I. S. (1988) Proc. Natl. Acad. Sci. USA 85, 4686-4690. 1. Haseltine, W. A. (1989) J. Acquired Immune Defic. Syndr. 2, 35. Gottlinger, H. G., Sodroski, J. G. & Haseltine, W. A. (1989) 311-344. Proc. Natl. Acad. Sci. USA 86, 5781-5785. 2. Cullen, B. R. & Greene, W. C. (1989) Cell 58, 423-426. 36. Kikuchi, Y., Ando, Y. & Shiba, T. (1986) Nature (London) 323, 3. Vaishnav, Y. N. & Wong-Staal, F. (1991) Annu. Rev. Biochem. 824-826. 60, 577-630. 37. Flavell, A. J. & Ish-Horowicz, D. (1983) Cell 34, 415-419. 4. Gelderblom, H. R. (1991) AIDS 5, 617-638. 38. Saigo, K. (1986) Nucleic Acids Res. 14, 4370-4371. 5. Hu, W. S. & Temin, H. M. (1990) Science 250, 1227-1233. 39. Olsen, J. & Swanstrom, R. (1985) J. Virol. 56, 779-789. 6. Varmus, H. & Swanstrom, R. (1985) RNA Tumor Viruses 2, 40. Sommer, R. & Tautz, D. (1989) Nucleic Acids Res. 17, 6749. 75-134. 41. Barklis, E., Mulligan, R. C. & Jaenisch, R. (1986) Cell 47, 7. Myers, G., Berzofsky, J. A., Rabson, A., Smith, T. F. & 391-399. Wong-Staal, F. (1990) Human Retroviruses and AIDS (Los 42. Loh, T. P., Sievert, L. L. & Scott, R. W. (1990) Mol. Cell. Alamos Natl. Lab., Los Alamos, NM). Biol. 10, 4045-4057. Downloaded by guest on September 24, 2021